talk, you’re on camera

06 and 07 aud vid comparison

library(tidyverse)
## Warning: package 'tidyverse' was built under R version 3.3.2
## Loading tidyverse: ggplot2
## Loading tidyverse: tibble
## Loading tidyverse: tidyr
## Loading tidyverse: readr
## Loading tidyverse: purrr
## Loading tidyverse: dplyr
## Warning: package 'ggplot2' was built under R version 3.3.2
## Warning: package 'tibble' was built under R version 3.3.2
## Warning: package 'tidyr' was built under R version 3.3.2
## Warning: package 'readr' was built under R version 3.3.2
## Warning: package 'purrr' was built under R version 3.3.2
## Conflicts with tidy packages ----------------------------------------------
## filter(): dplyr, stats
## lag():    dplyr, stats
library(feather)
## Warning: package 'feather' was built under R version 3.3.2
library(ggrepel)
## Warning: package 'ggrepel' was built under R version 3.3.2
library(broom)
## Warning: package 'broom' was built under R version 3.3.2
sixseven_basiclevel_home_data <- read_feather("data/all_basiclevel_home_data_feather5-17-17")%>%
  filter(month%in% c("06","07"))%>%
  droplevels()
  
sixseven_basiclevel_home_data_agg <- read_feather("data/all_basiclevel_home_data_agg_feather5-17-17")%>%
  filter(month%in% c("06","07"))%>%
  droplevels()
sixupdated <- sixseven_basiclevel_home_data_agg%>%
  filter(month=="06")
summary(sixseven_basiclevel_home_data_agg$subj, maxsum = 50)
## 01 02 03 04 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 27 
##  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4  3  4  4  4  4  4  4  4  4  4 
## 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 
##  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4  4
#we want a spread version of the data too
audio_ag <- sixseven_basiclevel_home_data_agg%>%
  filter(audio_video=="audio" & SubjectNumber!="17_06")%>%
  dplyr::select(-noun_chi_onset, -posttalk)%>%
  setNames(paste0('a_', names(.)))%>%
  rename(subj=a_subj,
         month = a_month,
         SubjectNumber= a_SubjectNumber)

sixseven_spreadAV <- sixseven_basiclevel_home_data_agg%>%
  filter(audio_video=="video")%>%
  dplyr::select(-noun_chi_onset, -posttalk)%>%
  setNames(paste0('v_', names(.)))%>%
  rename(subj=v_subj,
         month = v_month,
         SubjectNumber= v_SubjectNumber)%>%
  bind_cols(audio_ag)

#month spread
six_spreadmonth <- sixseven_basiclevel_home_data_agg%>%
  filter(month=="06")%>%
  dplyr::select(-noun_chi_onset, -posttalk, -SubjectNumber)%>%
  setNames(paste0('six_', names(.)))%>%
  rename(subj=six_subj,
         month = six_month,
         audio_video = six_audio_video)

sixseven_spreadmonth <- sixseven_basiclevel_home_data_agg%>%
  filter(month=="07" & !(audio_video=="video" & subj=="17"))%>%
  dplyr::select(-noun_chi_onset, -posttalk, -SubjectNumber)%>%
  setNames(paste0('sev_', names(.)))%>%
  rename(subj=sev_subj,
         month = sev_month,
         audio_video = sev_audio_video)%>%
  bind_cols(six_spreadmonth)

sheer #s of speakers, types, tokens, utterance types

#numspeakers
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, numspeakers, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#numtypes
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, numtokens, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#numtokens
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, numtypes, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#FAT huge diff, maybe 10fold?
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, FAT, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#MOT: big diff but not 10fold
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, MOT, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

# d big diff but not 10fold
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, i, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#q 
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, q, color=audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#s
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, s, color=audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#r
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, r, color=audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#n
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, n, color=audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#i
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, i, color=audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#TOY
ggplot(sixseven_basiclevel_home_data_agg, 
        aes(month, TOY, label=subj, color= audio_video))+
   geom_violin()+
   geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#TOY without 0s & with labels
ggplot(sixseven_basiclevel_home_data_agg %>%filter(TOY>0), 
        aes(month, TOY, label=subj, color= audio_video))+
   geom_violin()+
   #geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
   stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")+
     geom_label_repel(
  label.size = .10,
  label.padding = unit(0.1, "lines"))+scale_y_log10()

various props

#prop_op: higher for video by roughly 10%
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, prop_op, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#prop_mom: higher for video by roughly 15%
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, prop_mom, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#prop_dad: comparably low in both;~15%
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, prop_dad, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#prop_parent: same
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, prop_parent, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#prop_tech: same. identical.
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, prop_tech, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#propd: 10% more in audio (looks sig dif)
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, propd, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#propi: same
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, propi, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#propn: same
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, propn, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#propq: more in vid
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, propq, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#propr: very similar in both
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, propr, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#props: very similar in both
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, props, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#type token: 15% higher in video
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, type_token_ratio, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

#ent_subj_avg: identical
ggplot(sixseven_basiclevel_home_data_agg, 
       aes(month, ent_subj_av, color=audio_video))+
  geom_violin()+
  geom_point(position = position_jitterdodge(dodge.width=.9, jitter.width = .1, jitter.height = 0), shape = 21)+
  stat_summary(fun.data=mean_cl_boot, position = position_dodge(width=.9), aes(group=audio_video), color="black")

actual comparison across audio/video numerically

summary(sixseven_basiclevel_home_data_agg)
##       subj     month   SubjectNumber audio_video  numspeakers    
##  01     :  4   06:87   01_06  :  2   audio:88    Min.   : 1.000  
##  02     :  4   07:88   01_07  :  2   video:87    1st Qu.: 3.000  
##  03     :  4           02_06  :  2               Median : 4.000  
##  04     :  4           02_07  :  2               Mean   : 4.909  
##  06     :  4           03_06  :  2               3rd Qu.: 6.000  
##  07     :  4           03_07  :  2               Max.   :16.000  
##  (Other):151           (Other):163                               
##    numtokens         numtypes          FAT              MOT      
##  Min.   :  11.0   Min.   :  6.0   Min.   :  0.00   Min.   :   0  
##  1st Qu.: 141.0   1st Qu.: 50.5   1st Qu.:  0.00   1st Qu.:  98  
##  Median : 299.0   Median :103.0   Median :  8.00   Median : 205  
##  Mean   : 442.1   Mean   :121.3   Mean   : 61.15   Mean   : 276  
##  3rd Qu.: 677.0   3rd Qu.:186.0   3rd Qu.: 71.50   3rd Qu.: 381  
##  Max.   :1791.0   Max.   :379.0   Max.   :768.00   Max.   :1486  
##                                                                  
##  num_exp_tokens   num_exp_types          d               q         
##  Min.   :  0.00   Min.   : 0.000   Min.   :  5.0   Min.   :  0.00  
##  1st Qu.: 22.00   1st Qu.: 6.500   1st Qu.: 57.0   1st Qu.: 32.50  
##  Median : 57.00   Median :10.000   Median :130.0   Median : 69.00  
##  Mean   : 81.33   Mean   : 9.863   Mean   :206.6   Mean   : 89.48  
##  3rd Qu.:113.50   3rd Qu.:13.000   3rd Qu.:307.5   3rd Qu.:118.00  
##  Max.   :335.00   Max.   :17.000   Max.   :879.0   Max.   :469.00  
##                                                                    
##        s                r                n                i         
##  Min.   :  0.00   Min.   :  0.00   Min.   :  0.00   Min.   :  0.00  
##  1st Qu.:  3.00   1st Qu.:  0.00   1st Qu.:  9.50   1st Qu.:  7.50  
##  Median : 15.00   Median : 10.00   Median : 27.00   Median : 19.00  
##  Mean   : 44.61   Mean   : 36.19   Mean   : 38.36   Mean   : 26.82  
##  3rd Qu.: 41.00   3rd Qu.: 42.00   3rd Qu.: 51.00   3rd Qu.: 37.00  
##  Max.   :564.00   Max.   :256.00   Max.   :219.00   Max.   :173.00  
##                                                                     
##       TOY              n_op             y_op           prop_op      
##  Min.   :  0.00   Min.   :   7.0   Min.   :   3.0   Min.   :0.1053  
##  1st Qu.:  0.00   1st Qu.:  56.5   1st Qu.:  70.0   1st Qu.:0.3860  
##  Median :  0.00   Median : 122.0   Median : 145.0   Median :0.5240  
##  Mean   : 14.23   Mean   : 214.9   Mean   : 226.3   Mean   :0.5277  
##  3rd Qu.: 12.50   3rd Qu.: 327.0   3rd Qu.: 318.5   3rd Qu.:0.6782  
##  Max.   :383.00   Max.   :1096.0   Max.   :1046.0   Max.   :0.8992  
##                                                                     
##     n_op_exp         y_op_exp       prop_op_exp          CHI   
##  Min.   :  0.00   Min.   :  0.00   Min.   :0.0000   Min.   :0  
##  1st Qu.:  8.50   1st Qu.: 13.00   1st Qu.:0.4356   1st Qu.:0  
##  Median : 23.00   Median : 30.00   Median :0.5753   Median :0  
##  Mean   : 33.99   Mean   : 47.24   Mean   :0.5535   Mean   :0  
##  3rd Qu.: 51.00   3rd Qu.: 66.00   3rd Qu.:0.7040   3rd Qu.:0  
##  Max.   :175.00   Max.   :264.00   Max.   :0.9524   Max.   :0  
##                                                                
##     CHItypes    prop_mom         prop_dad        prop_parent    
##  Min.   :0   Min.   :0.0000   Min.   :0.00000   Min.   :0.0000  
##  1st Qu.:0   1st Qu.:0.4961   1st Qu.:0.00000   1st Qu.:0.7163  
##  Median :0   Median :0.7377   Median :0.02222   Median :0.8746  
##  Mean   :0   Mean   :0.6581   Mean   :0.13062   Mean   :0.7887  
##  3rd Qu.:0   3rd Qu.:0.8807   3rd Qu.:0.17247   3rd Qu.:0.9719  
##  Max.   :0   Max.   :1.0000   Max.   :1.00000   Max.   :1.0000  
##                                                                 
##    prop_tech             tech            propd             propi        
##  Min.   :0.000000   Min.   :  0.00   Min.   :0.07092   Min.   :0.00000  
##  1st Qu.:0.000000   1st Qu.:  0.00   1st Qu.:0.36767   1st Qu.:0.03149  
##  Median :0.007463   Median :  2.00   Median :0.46746   Median :0.05847  
##  Mean   :0.048776   Mean   : 26.46   Mean   :0.45019   Mean   :0.06651  
##  3rd Qu.:0.052437   3rd Qu.: 18.50   3rd Qu.:0.54669   3rd Qu.:0.08723  
##  Max.   :0.602837   Max.   :472.00   Max.   :0.73485   Max.   :0.22436  
##                                                                         
##      propn             propq            propr             props         
##  Min.   :0.00000   Min.   :0.0000   Min.   :0.00000   Min.   :0.000000  
##  1st Qu.:0.04956   1st Qu.:0.1683   1st Qu.:0.00000   1st Qu.:0.009472  
##  Median :0.08065   Median :0.2143   Median :0.02739   Median :0.035000  
##  Mean   :0.09466   Mean   :0.2282   Mean   :0.06887   Mean   :0.091467  
##  3rd Qu.:0.11601   3rd Qu.:0.2793   3rd Qu.:0.10614   3rd Qu.:0.120044  
##  Max.   :0.52941   Max.   :0.5870   Max.   :0.56647   Max.   :0.808511  
##                                                                         
##  type_token_ratio exp_type_ratio    exp_token_ratio   ent_subj_av   
##  Min.   :0.1435   Min.   :0.00000   Min.   :0.0000   Min.   :1.039  
##  1st Qu.:0.2644   1st Qu.:0.06760   1st Qu.:0.1425   1st Qu.:1.700  
##  Median :0.3289   Median :0.09244   Median :0.1782   Median :1.901  
##  Mean   :0.3297   Mean   :0.10721   Mean   :0.1850   Mean   :1.875  
##  3rd Qu.:0.3857   3rd Qu.:0.13423   3rd Qu.:0.2252   3rd Qu.:2.048  
##  Max.   :0.6441   Max.   :0.38462   Max.   :0.4824   Max.   :2.529  
##                                                                     
##   sum_prop_ut     noun_chi_onset   posttalk      
##  Min.   :0.9900   Min.   : 8.00   Mode :logical  
##  1st Qu.:1.0000   1st Qu.:10.00   FALSE:175      
##  Median :1.0000   Median :12.00   NA's :0        
##  Mean   :0.9999   Mean   :12.04                  
##  3rd Qu.:1.0000   3rd Qu.:14.00                  
##  Max.   :1.0000   Max.   :17.00                  
##                   NA's   :8
sixseven_basiclevel_home_data_agg%>%
  filter(subj!="17")%>%#missing a recording
  summarise_each(funs(wilcox.test(.[audio_video == "audio"], .[audio_video == "video"])$p.value,
                         cor.test(.[audio_video == "audio"], .[audio_video == "video"], method = "spearman")$p.value,
                      cor.test(.[audio_video == "audio"], .[audio_video == "video"], method = "spearman")$estimate),
                      #cor.test(.[audio_video == "audio"], .[audio_video == "video"], method = "spearman")$estimate),
                 vars = numspeakers:ent_subj_av)
## Warning in cor.test.default(c(4, 3, 5, 2, 7, 2, 7, 1, 4, 2, 6, 3, 12, 5, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(440, 120, 532, 114, 873, 58, 743, 49, 606, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(125, 62, 182, 31, 260, 26, 203, 24, 269,
## 29, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(89, 7, 195, 0, 69, 0, 3, 0, 47, 0, 44, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(327, 112, 317, 111, 701, 56, 542, 49, 495, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(154, 9, 66, 16, 134, 9, 158, 5, 71, 6, 78, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(12, 6, 14, 6, 15, 4, 14, 2, 13, 3, 12, 4, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(176, 47, 282, 26, 446, 29, 361, 19, 233,
## 23, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(64, 28, 79, 18, 188, 26, 148, 27, 88, 19, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(28, 2, 92, 68, 41, 0, 91, 0, 12, 6, 13, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(67, 29, 0, 0, 116, 0, 79, 0, 220, 0, 86,
## 15, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(84, 13, 57, 0, 42, 2, 24, 0, 36, 5, 32, 21, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(21, 1, 22, 2, 40, 1, 40, 3, 17, 1, 16, 4, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 17, 0, 0, 0, 0, 4, 383, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(189, 53, 328, 35, 381, 25, 277, 19, 284,
## 21, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(251, 67, 185, 79, 481, 33, 401, 30, 322,
## 33, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.570454545454545, 0.558333333333333,
## 0.360623781676413, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(57, 6, 37, 3, 49, 1, 39, 1, 37, 5, 14, 3, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(97, 3, 28, 13, 85, 8, 106, 4, 34, 1, 62,
## 33, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.62987012987013, 0.333333333333333,
## 0.430769230769231, : Cannot compute exact p-value with ties
## Warning in cor(rank(x), rank(y)): the standard deviation is zero

## Warning in cor(rank(x), rank(y)): the standard deviation is zero
## Warning in cor.test.default(c(0.743181818181818, 0.933333333333333,
## 0.595864661654135, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.202272727272727, 0.0583333333333333,
## 0.366541353383459, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.945454545454545, 0.991666666666667,
## 0.962406015037594, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 0.0228802153432032, 0,
## 0.0066006600660066, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 17, 0, 4, 0, 0, 4, 390, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.4, 0.391666666666667, 0.530075187969925, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.0477272727272727, 0.00833333333333333,
## 0.0413533834586466, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.190909090909091, 0.108333333333333,
## 0.107142857142857, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.145454545454545, 0.233333333333333,
## 0.148496240601504, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.152272727272727, 0.241666666666667, 0, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.0636363636363636, 0.0166666666666667,
## 0.172932330827068, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.284090909090909, 0.516666666666667,
## 0.342105263157895, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.096, 0.0967741935483871,
## 0.0769230769230769, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(4, 3, 5, 2, 7, 2, 7, 1, 4, 2, 6, 3, 12, 5, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(440, 120, 532, 114, 873, 58, 743, 49, 606, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(125, 62, 182, 31, 260, 26, 203, 24, 269,
## 29, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(89, 7, 195, 0, 69, 0, 3, 0, 47, 0, 44, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(327, 112, 317, 111, 701, 56, 542, 49, 495, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(154, 9, 66, 16, 134, 9, 158, 5, 71, 6, 78, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(12, 6, 14, 6, 15, 4, 14, 2, 13, 3, 12, 4, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(176, 47, 282, 26, 446, 29, 361, 19, 233,
## 23, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(64, 28, 79, 18, 188, 26, 148, 27, 88, 19, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(28, 2, 92, 68, 41, 0, 91, 0, 12, 6, 13, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(67, 29, 0, 0, 116, 0, 79, 0, 220, 0, 86,
## 15, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(84, 13, 57, 0, 42, 2, 24, 0, 36, 5, 32, 21, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(21, 1, 22, 2, 40, 1, 40, 3, 17, 1, 16, 4, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 17, 0, 0, 0, 0, 4, 383, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(189, 53, 328, 35, 381, 25, 277, 19, 284,
## 21, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(251, 67, 185, 79, 481, 33, 401, 30, 322,
## 33, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.570454545454545, 0.558333333333333,
## 0.360623781676413, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(57, 6, 37, 3, 49, 1, 39, 1, 37, 5, 14, 3, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(97, 3, 28, 13, 85, 8, 106, 4, 34, 1, 62,
## 33, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.62987012987013, 0.333333333333333,
## 0.430769230769231, : Cannot compute exact p-value with ties
## Warning in cor(rank(x), rank(y)): the standard deviation is zero

## Warning in cor(rank(x), rank(y)): the standard deviation is zero
## Warning in cor.test.default(c(0.743181818181818, 0.933333333333333,
## 0.595864661654135, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.202272727272727, 0.0583333333333333,
## 0.366541353383459, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.945454545454545, 0.991666666666667,
## 0.962406015037594, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 0.0228802153432032, 0,
## 0.0066006600660066, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0, 0, 0, 0, 0, 0, 17, 0, 4, 0, 0, 4, 390, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.4, 0.391666666666667, 0.530075187969925, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.0477272727272727, 0.00833333333333333,
## 0.0413533834586466, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.190909090909091, 0.108333333333333,
## 0.107142857142857, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.145454545454545, 0.233333333333333,
## 0.148496240601504, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.152272727272727, 0.241666666666667, 0, 0, :
## Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.0636363636363636, 0.0166666666666667,
## 0.172932330827068, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.284090909090909, 0.516666666666667,
## 0.342105263157895, : Cannot compute exact p-value with ties
## Warning in cor.test.default(c(0.096, 0.0967741935483871,
## 0.0769230769230769, : Cannot compute exact p-value with ties
## # A tibble: 1 x 37
##   `vars1_$` `vars2_$` `vars3_$` `vars4_$` `vars5_$` `vars6_$` `vars7_$`
##       <dbl>     <dbl>     <dbl>     <dbl>     <dbl>     <dbl>     <dbl>
## 1 0.1794435 0.4897253 0.4419861 0.3550851 0.4923028 0.4470059 0.4089341
## # ... with 30 more variables: `vars8_$` <dbl>, `vars9_$` <dbl>,
## #   `vars10_$` <dbl>, `vars11_$` <dbl>, `vars12_$` <dbl>,
## #   `vars13_$` <dbl>, `vars14_$` <dbl>, `vars15_$` <dbl>,
## #   `vars16_$` <dbl>, `vars17_$` <dbl>, `vars18_$` <dbl>,
## #   `vars19_$` <dbl>, `vars20_$` <dbl>, `vars21_$` <dbl>,
## #   `vars22_$` <dbl>, `vars23_$` <dbl>, `vars24_$` <dbl>,
## #   `vars25_$` <dbl>, `vars26_$` <dbl>, `vars27_$` <dbl>,
## #   `vars28_$` <dbl>, `vars29_$` <dbl>, `vars30_$` <dbl>,
## #   `vars31_$` <dbl>, `vars32_$` <dbl>, `vars33_$` <dbl>,
## #   `vars34_$` <dbl>, `vars35_$` <dbl>, `vars36_$` <dbl>, `vars37_$` <dbl>
#View(sixseven_basiclevel_home_data_agg)

correlation graphs

ggplot(sixseven_spreadAV,
       aes(a_numspeakers, v_numspeakers, color=month, fill = month))+
  geom_point()+
  stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#corr
ggplot(sixseven_spreadAV,
       aes(a_numtypes, v_numtypes, color=month, fill = month))+
  geom_point()+
  stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_numtokens, v_numtokens, color=month, fill = month))+
  geom_point()+
  stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_FAT, v_FAT, color=month, fill = month))+
  geom_point()+
  stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#corr
ggplot(sixseven_spreadAV,
       aes(a_MOT, v_MOT, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

colnames(sixseven_spreadAV)
##  [1] "subj"               "month"              "SubjectNumber"     
##  [4] "v_audio_video"      "v_numspeakers"      "v_numtokens"       
##  [7] "v_numtypes"         "v_FAT"              "v_MOT"             
## [10] "v_num_exp_tokens"   "v_num_exp_types"    "v_d"               
## [13] "v_q"                "v_s"                "v_r"               
## [16] "v_n"                "v_i"                "v_TOY"             
## [19] "v_n_op"             "v_y_op"             "v_prop_op"         
## [22] "v_n_op_exp"         "v_y_op_exp"         "v_prop_op_exp"     
## [25] "v_CHI"              "v_CHItypes"         "v_prop_mom"        
## [28] "v_prop_dad"         "v_prop_parent"      "v_prop_tech"       
## [31] "v_tech"             "v_propd"            "v_propi"           
## [34] "v_propn"            "v_propq"            "v_propr"           
## [37] "v_props"            "v_type_token_ratio" "v_exp_type_ratio"  
## [40] "v_exp_token_ratio"  "v_ent_subj_av"      "v_sum_prop_ut"     
## [43] "subj"               "month"              "SubjectNumber"     
## [46] "a_audio_video"      "a_numspeakers"      "a_numtokens"       
## [49] "a_numtypes"         "a_FAT"              "a_MOT"             
## [52] "a_num_exp_tokens"   "a_num_exp_types"    "a_d"               
## [55] "a_q"                "a_s"                "a_r"               
## [58] "a_n"                "a_i"                "a_TOY"             
## [61] "a_n_op"             "a_y_op"             "a_prop_op"         
## [64] "a_n_op_exp"         "a_y_op_exp"         "a_prop_op_exp"     
## [67] "a_CHI"              "a_CHItypes"         "a_prop_mom"        
## [70] "a_prop_dad"         "a_prop_parent"      "a_prop_tech"       
## [73] "a_tech"             "a_propd"            "a_propi"           
## [76] "a_propn"            "a_propq"            "a_propr"           
## [79] "a_props"            "a_type_token_ratio" "a_exp_type_ratio"  
## [82] "a_exp_token_ratio"  "a_ent_subj_av"      "a_sum_prop_ut"
#corr
ggplot(sixseven_spreadAV,
       aes(a_d, v_d, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#corr
ggplot(sixseven_spreadAV,
       aes(a_q, v_q, color=month, fill = month))+
  geom_point()+
    stat_smooth(method="rlm")+facet_wrap(~month)
## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

#
ggplot(sixseven_spreadAV,
       aes(a_i, v_i, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_n, v_n, color=month, fill = month))+
  geom_point()+
    stat_smooth(method="rlm")+facet_wrap(~month)
## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

ggplot(sixseven_spreadAV,
       aes(a_r, v_r, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_s, v_s, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#crazy spread
ggplot(sixseven_spreadAV,
       aes(a_TOY, v_TOY, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#not corr
ggplot(sixseven_spreadAV,
       aes(a_prop_op, v_prop_op, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#corr
ggplot(sixseven_spreadAV,
       aes(a_prop_mom, v_prop_mom, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#sparse
ggplot(sixseven_spreadAV,
       aes(a_prop_dad, v_prop_dad, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#weird dist
ggplot(sixseven_spreadAV,
       aes(a_prop_parent, v_prop_parent, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#sparse
ggplot(sixseven_spreadAV,
       aes(a_prop_tech, v_prop_tech, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_propd, v_propd, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_propq, v_propq, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_props, v_props, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

#corr
ggplot(sixseven_spreadAV,
       aes(a_propi, v_propi, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_propn, v_propn, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_propr, v_propr, color=month, fill = month))+
  geom_point()+
    stat_smooth()+facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadAV,
       aes(a_type_token_ratio, v_type_token_ratio, color=month, fill = month))+
  geom_point()+
    stat_smooth(method="rlm")+facet_wrap(~month)
## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

shapiro.test(sixseven_spreadAV$v_type_token_ratio)
## 
##  Shapiro-Wilk normality test
## 
## data:  sixseven_spreadAV$v_type_token_ratio
## W = 0.98309, p-value = 0.3154
ggplot(sixseven_spreadAV,
       aes(a_ent_subj_av, v_ent_subj_av, color=month, fill = month))+
  geom_point()+
    stat_smooth(method="rlm")+facet_wrap(~month)
## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

distribution of utterance_types

#ut stacked proportionally
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = utterance_type))+
  geom_bar(position = "fill")+
  facet_wrap(~month)

#ut stacked by count
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = utterance_type))+
  geom_bar()+
  facet_wrap(~month)

#op stacked proportionally
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = object_present))+
  geom_bar(position = "fill")+
  facet_wrap(~month)

#op stacked by count
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = object_present))+
  geom_bar()+
  facet_wrap(~month)

#speaker stacked proportionally
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = speaker))+
  geom_bar(position = "fill")+
  facet_wrap(~month)

#speaker stacked by count
ggplot(sixseven_basiclevel_home_data, 
       aes(audio_video, fill = speaker))+
  geom_bar()+
  facet_wrap(~month)

top100av <- sixseven_basiclevel_home_data%>%
  group_by(audio_video, basic_level)%>%
  tally()%>%
  top_n(100,n)

top100a <- filter(top100av, audio_video=="audio")%>%rename(naud=n)
top100v <- filter(top100av, audio_video=="video")%>%rename(nvid=n)

table(top100a$basic_level %in% top100v$basic_level)
## 
## FALSE  TRUE 
##    26    74
table(top100v$basic_level %in% top100a$basic_level)
## 
## FALSE  TRUE 
##    26    74
top100a%>%
  filter(basic_level %in% top100v$basic_level)%>%
  arrange(-naud)
## Source: local data frame [74 x 3]
## Groups: audio_video [1]
## 
## # A tibble: 74 x 3
##    audio_video basic_level  naud
##         <fctr>      <fctr> <int>
##  1       audio        hand  1251
##  2       audio        baby  1195
##  3       audio      diaper  1135
##  4       audio        foot  1133
##  5       audio        book  1055
##  6       audio         toe   825
##  7       audio      bottle   812
##  8       audio        hair   794
##  9       audio        head   687
## 10       audio        face   672
## # ... with 64 more rows
top100v%>%
  filter(basic_level %in% top100a$basic_level)%>%
  arrange(-nvid)
## Source: local data frame [74 x 3]
## Groups: audio_video [1]
## 
## # A tibble: 74 x 3
##    audio_video basic_level  nvid
##         <fctr>      <fctr> <int>
##  1       video         hat   453
##  2       video        book   370
##  3       video        ball   334
##  4       video        baby   281
##  5       video        hand   227
##  6       video       doggy   215
##  7       video         dog   193
##  8       video        foot   190
##  9       video       mouth   190
## 10       video         toe   181
## # ... with 64 more rows
left_join(top100a, top100v)
## Joining, by = c("audio_video", "basic_level")
## Source: local data frame [100 x 4]
## Groups: audio_video [?]
## 
## # A tibble: 100 x 4
##    audio_video basic_level  naud  nvid
##         <fctr>      <fctr> <int> <int>
##  1       audio       apple   143    NA
##  2       audio      apples   231    NA
##  3       audio         arm   478    NA
##  4       audio        baby  1195    NA
##  5       audio        back   133    NA
##  6       audio         bag   317    NA
##  7       audio        ball   606    NA
##  8       audio      banana   214    NA
##  9       audio     bananas   275    NA
## 10       audio        bath   285    NA
## # ... with 90 more rows
hist_six <- sixseven_basiclevel_home_data%>%
  filter(month=="06")%>%
  group_by(subj, audio_video)%>%
  tally()

ggplot(sixseven_basiclevel_home_data%>%filter(month=="06"), aes(subj, fill = audio_video))+
         geom_bar()+facet_wrap(~audio_video)

ggplot(hist_six, aes(n, fill=audio_video))+geom_histogram(binwidth=100)+facet_wrap(~audio_video, scales ="free")

ggplot(sixseven_spreadmonth, aes(six_numspeakers, sev_numspeakers))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at 0.975
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.025
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 1.7362e-16
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 1
## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at 0.975
## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.025
## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 1.7362e-16
## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 1

ggplot(sixseven_spreadmonth, aes(six_numtypes, sev_numtypes))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_FAT, sev_FAT))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_MOT, sev_MOT))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_d, sev_d))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_q, sev_q))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_i, sev_i))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_r, sev_r))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_s, sev_s))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_n, sev_n))+geom_point()+stat_smooth()+facet_wrap(~audio_video, scales = "free", nrow  =2)
## `geom_smooth()` using method = 'loess'

stats

#need to redo as moving window
lm_quant_prop <- lm(data = sixseven_basiclevel_home_data_agg, type_token_ratio ~ month + audio_video)
anova(lm_quant_prop)
## Analysis of Variance Table
## 
## Response: type_token_ratio
##              Df  Sum Sq  Mean Sq F value    Pr(>F)    
## month         1 0.00088 0.000884  0.1269    0.7221    
## audio_video   1 0.27410 0.274099 39.3400 2.788e-09 ***
## Residuals   172 1.19840 0.006967                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
plot(lm_quant_prop)

summary(lm_quant_prop)
## 
## Call:
## lm(formula = type_token_ratio ~ month + audio_video, data = sixseven_basiclevel_home_data_agg)
## 
## Residuals:
##       Min        1Q    Median        3Q       Max 
## -0.184288 -0.058783 -0.008734  0.049558  0.277039 
## 
## Coefficients:
##                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       0.292825   0.010908  26.844  < 2e-16 ***
## month07          -0.004951   0.012620  -0.392    0.695    
## audio_videovideo  0.079155   0.012620   6.272 2.79e-09 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.08347 on 172 degrees of freedom
## Multiple R-squared:  0.1866, Adjusted R-squared:  0.1772 
## F-statistic: 19.73 on 2 and 172 DF,  p-value: 1.926e-08
sixseven_basiclevel_home_data_agg %>% 
  group_by(month, audio_video)%>%
  summarise(mean(type_token_ratio))
## Source: local data frame [4 x 3]
## Groups: month [?]
## 
## # A tibble: 4 x 3
##    month audio_video `mean(type_token_ratio)`
##   <fctr>      <fctr>                    <dbl>
## 1     06       audio                0.2903638
## 2     06       video                0.3744989
## 3     07       audio                0.2903356
## 4     07       video                0.3645678
#sixseven_spreadAV %>% 
sixseven_basiclevel_home_data_agg %>% 
  group_by(month, audio_video)%>%
  summarise(median(numspeakers, na.rm=T),
            mean(numspeakers, na.rm=T),
            sd(numspeakers, na.rm=T))
## Source: local data frame [4 x 5]
## Groups: month [?]
## 
## # A tibble: 4 x 5
##    month audio_video `median(numspeakers, na.rm = T)`
##   <fctr>      <fctr>                            <dbl>
## 1     06       audio                                5
## 2     06       video                                3
## 3     07       audio                                7
## 4     07       video                                3
## # ... with 2 more variables: `mean(numspeakers, na.rm = T)` <dbl>,
## #   `sd(numspeakers, na.rm = T)` <dbl>
sixseven_basiclevel_home_data_agg %>% 
  group_by(month, audio_video)%>%
  summarise(median(FAT, na.rm=T),
            mean(FAT, na.rm=T),
            sd(FAT, na.rm=T))
## Source: local data frame [4 x 5]
## Groups: month [?]
## 
## # A tibble: 4 x 5
##    month audio_video `median(FAT, na.rm = T)` `mean(FAT, na.rm = T)`
##   <fctr>      <fctr>                    <dbl>                  <dbl>
## 1     06       audio                     67.0             103.250000
## 2     06       video                      0.0              15.837209
## 3     07       audio                     40.5             115.431818
## 4     07       video                      0.0               9.045455
## # ... with 1 more variables: `sd(FAT, na.rm = T)` <dbl>
# none of the audio differences 6 vs 7 are different
wa1 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numtokens,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numtokens, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa2 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numtypes,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numtypes, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa3 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numspeakers,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numspeakers, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wa4 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$MOT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$MOT, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa5 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$FAT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$FAT, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wa6 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$d,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$d, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa7 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$q,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$q, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa8 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$r,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$r, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wa9 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$i,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$i, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wa10 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$n,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$n, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wa11 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$s,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$s, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wa12 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$y_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$n_op, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
#most video 6vs7 aren't different
wv1 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numtokens,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numtokens, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wv2 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numtypes,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numtypes, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wv3 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numspeakers,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numspeakers, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv4 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$MOT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$MOT, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv5 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$d,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$d, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wv6 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$r,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$r, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv7 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$i,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$i, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv8 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$y_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$y_op, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv9 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$q,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$q, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv10 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$n,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$n, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
wv11 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$s,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$s, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
wv12 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06" & subj!="17")$FAT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07" & subj!="17")$FAT, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with ties
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
# none of the audio differences 6 vs 7 are different
ca1 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numtokens,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numtokens, conf.int=T, method = "kendall")
ca2 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numtypes,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numtypes, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca3 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$numspeakers,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$numspeakers, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca4 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$MOT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$MOT, conf.int=T, method = "kendall")
ca5 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$FAT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$FAT, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca6 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$d,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$d, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca7 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$q,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$q, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca8 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$r,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$r, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca9 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$i,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$i, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca10 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$n,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$n, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca11 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$s,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$s, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
ca12 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$y_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$n_op, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
#most video 6vs7 aren't different
cv1 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numtokens,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numtokens, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv2 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numtypes,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numtypes, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv3 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$numspeakers,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$numspeakers, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv4 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$MOT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$MOT, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv5 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$d,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$d, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv6 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$r,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$r, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv7 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$i,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$i, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv8 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$y_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$y_op, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv9 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$q,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$q, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv10 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$n,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$n, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv11 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$s,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$s, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv12 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06" & subj!="17")$FAT,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07" & subj!="17")$FAT, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
cv_ps <- cbind(cv1$p.value,cv2$p.value,cv3$p.value, cv4$p.value, cv5$p.value, cv6$p.value, cv7$p.value, cv8$p.value, cv9$p.value, cv10$p.value, cv11$p.value, cv12$p.value)
round(p.adjust(cv_ps, method = "holm"),3)
##  [1] 0.000 0.000 0.014 0.000 0.000 0.066 0.004 0.001 0.000 0.001 0.024
## [12] 0.000
round(cv_ps,3)
##      [,1] [,2]  [,3] [,4] [,5]  [,6]  [,7] [,8] [,9] [,10] [,11] [,12]
## [1,]    0    0 0.005    0    0 0.066 0.001    0    0     0 0.012     0
table(cv_ps<.05)
## 
## FALSE  TRUE 
##     1    11
table(p.adjust(cv_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     1    11
cv6
## 
##  Kendall's rank correlation tau
## 
## data:  subset(sixseven_basiclevel_home_data_agg, audio_video == "video" &  and subset(sixseven_basiclevel_home_data_agg, audio_video == "video" &     month == "06" & subj != "17")$r and     month == "07" & subj != "17")$r
## z = 1.8384, p-value = 0.066
## alternative hypothesis: true tau is not equal to 0
## sample estimates:
##      tau 
## 0.229212
#taking out cv7 bc it's not sig, reading count
cv_taus <- cbind(cv1$estimate,cv2$estimate,cv3$estimate, cv4$estimate, cv5$estimate, cv6$estimate, cv8$estimate, cv9$estimate, cv10$estimate, cv11$estimate, cv12$estimate)
range(cv_taus)
## [1] 0.2292120 0.5973362
ca_ps <- cbind(ca1$p.value,ca2$p.value,ca3$p.value, ca4$p.value, ca5$p.value, ca6$p.value, ca7$p.value, ca8$p.value, ca9$p.value, ca10$p.value, ca11$p.value, ca12$p.value)
round(p.adjust(ca_ps, method = "holm"),3)
##  [1] 0.001 0.010 0.011 0.000 0.004 0.000 0.000 0.001 0.010 0.015 0.003
## [12] 0.006
round(ca_ps, 3)
##      [,1]  [,2]  [,3] [,4]  [,5] [,6] [,7] [,8]  [,9] [,10] [,11] [,12]
## [1,]    0 0.002 0.006    0 0.001    0    0    0 0.003 0.015     0 0.001
table(ca_ps<.05)
## 
## TRUE 
##   12
table(p.adjust(ca_ps, method = "holm")<.05)
## 
## TRUE 
##   12
wv_ps <- cbind(wv1$p.value,wv2$p.value,wv3$p.value, wv4$p.value, wv5$p.value, wv6$p.value, wv7$p.value, wv8$p.value, wv9$p.value, wv10$p.value, wv11$p.value, wv12$p.value)
round(p.adjust(wv_ps, method = "holm"),3)
##  [1] 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.420 0.292 0.199
## [12] 0.155
round(wv_ps, 3)
##       [,1] [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10] [,11]
## [1,] 0.534 0.62 0.163 0.591 0.249 0.516 0.692 0.557 0.047 0.029 0.018
##      [,12]
## [1,] 0.013
table(wv_ps<.05)
## 
## FALSE  TRUE 
##     8     4
table(p.adjust(wv_ps, method = "holm")<.05)
## 
## FALSE 
##    12
ca_taus <- cbind(ca1$estimate,ca2$estimate,ca3$estimate, ca4$estimate, ca5$estimate, ca6$estimate, ca7$estimate, ca8$estimate, ca9$estimate, ca10$estimate, ca11$estimate, ca12$estimate)
range(ca_taus)
## [1] 0.2565195 0.5137421
wa_ps <- cbind(wa1$p.value,wa2$p.value,wa3$p.value, wa4$p.value, wa5$p.value, wa6$p.value, wa7$p.value, wa8$p.value, wa9$p.value, wa10$p.value, wa11$p.value, wa12$p.value)
round(p.adjust(wa_ps, method = "holm"),3)
##  [1] 1 1 1 1 1 1 1 1 1 1 1 1
round(wa_ps, 3)
##       [,1]  [,2] [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10] [,11]
## [1,] 0.875 0.793 0.09 0.649 0.876 0.793 0.283 0.894 0.876 0.744 0.518
##      [,12]
## [1,] 0.649
table(wa_ps<.05)
## 
## FALSE 
##    12
table(p.adjust(wa_ps, method = "holm")<.05)
## 
## FALSE 
##    12
sixseven_basiclevel_home_data_agg%>%
  group_by(audio_video) %>% 
  summarise(meantokens = mean(numtokens, na.rm=T),
    meantokens_hour = mean(as.numeric(as.character(ifelse(audio_video=="audio", numtokens/11, numtokens))), na.rm=T))
## # A tibble: 2 x 3
##   audio_video meantokens meantokens_hour
##        <fctr>      <dbl>           <dbl>
## 1       audio   713.6023        64.87293
## 2       video   167.4023       167.40230
# none of the audio differences 6 vs 7 are different
pwa1 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_op, conf.int=T, paired = T)
pwa2 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_mom,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_mom, conf.int=T, paired = T)
pwa3 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_dad,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_dad, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwa4 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propd,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propd, conf.int=T, paired = T)
pwa5 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propi,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propi, conf.int=T, paired = T)
pwa6 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propr,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propr, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwa7 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propq,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propq, conf.int=T, paired = T)
pwa8 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$props,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$props, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwa9 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propn,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propn, conf.int=T, paired = T)
pwa10 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$type_token_ratio,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$type_token_ratio, conf.int=T, paired = T)

#most video 6vs7 aren't different
pwv1 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_op, conf.int=T, paired = T)
pwv2 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_mom,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_mom, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwv3 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_dad,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_dad, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwv4 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propd,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propd, conf.int=T, paired = T)
pwv5 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propi,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propi, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwv6 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propr,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propr, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwv7 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propq,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propq, conf.int=T, paired = T)
pwv8 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$props,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$props, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
pwv9 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propn,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propn, conf.int=T, paired = T)
pwv10 <- wilcox.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$type_token_ratio,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$type_token_ratio, conf.int=T, paired = T)
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(subset(sixseven_basiclevel_home_data_agg, :
## cannot compute exact confidence interval with zeroes
# none of the audio differences 6 vs 7 are different
pca1 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_op, conf.int=T, method = "kendall")
pca2 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_mom,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_mom, conf.int=T, method = "kendall")
pca3 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$prop_dad,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$prop_dad, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pca4 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propd,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propd, conf.int=T, method = "kendall")
pca5 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propi,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propi, conf.int=T, method = "kendall")
pca6 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propr,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propr, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pca7 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propq,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propq, conf.int=T, method = "kendall")
pca8 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$props,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$props, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pca9 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$propn,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$propn, conf.int=T, method = "kendall")
pca10 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="06")$type_token_ratio,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="audio" & month=="07")$type_token_ratio, conf.int=T, method = "kendall")

#most video 6vs7 aren't different
pcv1 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_op,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_op, conf.int=T, method = "kendall")
pcv2 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_mom,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_mom, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv3 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$prop_dad,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$prop_dad, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv4 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propd,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propd, conf.int=T, method = "kendall")
pcv5 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propi,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propi, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv6 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propr,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propr, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv7 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propq,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propq, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv8 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$props,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$props, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv9 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$propn,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$propn, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv10 <- cor.test(subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="06"& subj!="17")$type_token_ratio,
            subset(sixseven_basiclevel_home_data_agg, audio_video=="video" & month=="07"& subj!="17")$type_token_ratio, conf.int=T, method = "kendall")
## Warning in cor.test.default(subset(sixseven_basiclevel_home_data_agg,
## audio_video == : Cannot compute exact p-value with ties
pcv_ps <- cbind(pcv1$p.value,pcv2$p.value,pcv3$p.value, pcv4$p.value, pcv5$p.value, pcv6$p.value, pcv7$p.value, pcv8$p.value, pcv9$p.value, pcv10$p.value)
round(p.adjust(pcv_ps, method = "holm"),3)#prop_mom, prop_dad, 
##  [1] 0.713 0.000 0.000 0.293 0.069 0.713 0.713 0.713 0.713 0.713
round(pcv_ps,3)
##       [,1] [,2] [,3]  [,4]  [,5] [,6] [,7]  [,8]  [,9] [,10]
## [1,] 0.662    0    0 0.042 0.009 0.14 0.25 0.226 0.119 0.161
table(pcv_ps<.05)
## 
## FALSE  TRUE 
##     6     4
table(p.adjust(pcv_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     8     2
pcv2 %>% tidy()
##    estimate statistic      p.value                         method
## 1 0.4429467  4.071858 4.663969e-05 Kendall's rank correlation tau
##   alternative
## 1   two.sided
pcv3 %>% tidy()
##    estimate statistic      p.value                         method
## 1 0.6292412  4.811621 1.497114e-06 Kendall's rank correlation tau
##   alternative
## 1   two.sided
pca_ps <- cbind(pca1$p.value,pca2$p.value,pca3$p.value, pca4$p.value, pca5$p.value, pca6$p.value, pca7$p.value, pca8$p.value, pca9$p.value, pca10$p.value)
round(p.adjust(pca_ps, method = "holm"),3)# 2,3,4,5,6,8,10: all but propop, propq, and propn
##  [1] 0.487 0.017 0.007 0.007 0.000 0.007 0.216 0.017 0.060 0.000
round(pca_ps, 3)
##       [,1]  [,2]  [,3]  [,4] [,5]  [,6]  [,7]  [,8] [,9] [,10]
## [1,] 0.487 0.004 0.001 0.001    0 0.001 0.108 0.003 0.02     0
table(pca_ps<.05)
## 
## FALSE  TRUE 
##     2     8
table(p.adjust(pca_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     3     7
#removing 1 7 and 9 bc ns
pca_taus <- cbind(pca2$estimate,pca3$estimate, pca4$estimate, pca5$estimate, pca6$estimate, pca8$estimate, pca10$estimate)
range(pca_taus)
## [1] 0.2959831 0.4608879
pwv_ps <- cbind(pwv1$p.value,pwv2$p.value,pwv3$p.value, pwv4$p.value, pwv5$p.value, pwv6$p.value, pwv7$p.value, pwv8$p.value, pwv9$p.value, pwv10$p.value)
round(p.adjust(pwv_ps, method = "holm"),3)
##  [1] 1.000 1.000 0.535 1.000 1.000 1.000 0.535 0.161 1.000 1.000
round(pwv_ps, 3)
##       [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10]
## [1,] 0.948 0.223 0.059 0.611 0.395 0.641 0.063 0.016 0.223 0.832
table(pwv_ps<.05)
## 
## FALSE  TRUE 
##     9     1
table(p.adjust(pwv_ps, method = "holm")<.05)
## 
## FALSE 
##    10
# none of them differ

pwa_ps <- cbind(pwa1$p.value,pwa2$p.value,pwa3$p.value, pwa4$p.value, pwa5$p.value, pwa6$p.value, pwa7$p.value, pwa8$p.value, pwa9$p.value, pwa10$p.value)
round(p.adjust(pwa_ps, method = "holm"),3)
##  [1] 1 1 1 1 1 1 1 1 1 1
round(pwa_ps, 3)
##       [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10]
## [1,] 0.575 0.234 0.767 0.448 0.138 0.964 0.583 0.429 0.913     1
table(pwa_ps<.05)
## 
## FALSE 
##    10
table(p.adjust(pwa_ps, method = "holm")<.05)
## 
## FALSE 
##    10
#most six props aren't different
propwsix1 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), prop_op~audio_video, conf.int=T, paired = T)
propwsix2 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), prop_mom~audio_video, conf.int=T, paired = T)
propwsix3 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), prop_dad~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.202272727272727,
## 0.0790378006872852, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.202272727272727,
## 0.0790378006872852, : cannot compute exact confidence interval with zeroes
propwsix4 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), propd~audio_video, conf.int=T, paired = T)
propwsix5 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), propi~audio_video, conf.int=T, paired = T)
propwsix6 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), propr~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.152272727272727,
## 0.132875143184422, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.152272727272727,
## 0.132875143184422, : cannot compute exact confidence interval with zeroes
propwsix7 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), propq~audio_video, conf.int=T, paired = T)
propwsix8 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), props~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.0636363636363636,
## 0.0469644902634593, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.0636363636363636,
## 0.0469644902634593, : cannot compute exact confidence interval with zeroes
propwsix9 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), propn~audio_video, conf.int=T, paired = T)
propwsix10 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="06"& subj!="17"), type_token_ratio~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.284090909090909,
## 0.297823596792669, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.284090909090909,
## 0.297823596792669, : cannot compute exact confidence interval with zeroes
propwsix_ps <- cbind(propwsix1$p.value,propwsix2$p.value,propwsix3$p.value, propwsix4$p.value, propwsix5$p.value, propwsix6$p.value, propwsix7$p.value, propwsix8$p.value, propwsix9$p.value, propwsix10$p.value)
round(p.adjust(propwsix_ps, method = "holm"),3) #1, 4,7, 10: prop_op, propd, propq, and ttr
##  [1] 0.017 0.122 0.122 0.005 0.599 0.941 0.000 0.463 0.463 0.001
round(propwsix_ps, 3)
##       [,1] [,2]  [,3]  [,4] [,5]  [,6] [,7]  [,8]  [,9] [,10]
## [1,] 0.002 0.02 0.021 0.001  0.3 0.941    0 0.127 0.116     0
table(propwsix_ps<.05)
## 
## FALSE  TRUE 
##     4     6
table(p.adjust(propwsix_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     6     4
#most six props aren't different
propwseven1 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), prop_op~audio_video, conf.int=T, paired = T)
propwseven2 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), prop_mom~audio_video, conf.int=T, paired = T)
propwseven3 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), prop_dad~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.366541353383459,
## 0.00403768506056528, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.366541353383459,
## 0.00403768506056528, : cannot compute exact confidence interval with zeroes
propwseven4 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), propd~audio_video, conf.int=T, paired = T)
propwseven5 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), propi~audio_video, conf.int=T, paired = T)
propwseven6 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), propr~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0, 0.106325706594886,
## 0.192393736017897, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0, 0.106325706594886,
## 0.192393736017897, : cannot compute exact confidence interval with zeroes
propwseven7 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), propq~audio_video, conf.int=T, paired = T)
propwseven8 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), props~audio_video, conf.int=T, paired = T)
## Warning in wilcox.test.default(x = c(0.172932330827068,
## 0.122476446837147, : cannot compute exact p-value with zeroes
## Warning in wilcox.test.default(x = c(0.172932330827068,
## 0.122476446837147, : cannot compute exact confidence interval with zeroes
propwseven9 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), propn~audio_video, conf.int=T, paired = T)
propwseven10 <- wilcox.test(data = subset(sixseven_basiclevel_home_data_agg, month=="07"), type_token_ratio~audio_video, conf.int=T, paired = T)

propwseven_ps <- cbind(propwseven1$p.value,propwseven2$p.value,propwseven3$p.value, propwseven4$p.value, propwseven5$p.value, propwseven6$p.value, propwseven7$p.value, propwseven8$p.value, propwseven9$p.value, propwseven10$p.value)
round(p.adjust(propwseven_ps, method = "holm"),3) #1, 2,3,4,7,10: prop_op, prop_mom, prop_dad, propd, propq, and ttr
##  [1] 0.035 0.007 0.014 0.000 0.503 1.000 0.035 1.000 1.000 0.000
round(propwseven_ps, 3)
##       [,1]  [,2]  [,3] [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10]
## [1,] 0.007 0.001 0.002    0 0.126 0.338 0.006 0.368 0.977     0
table(propwseven_ps<.05)
## 
## FALSE  TRUE 
##     4     6
table(p.adjust(propwseven_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     4     6
propwseven_estdiff <- cbind(propwseven1$estimate,propwseven2$estimate,propwseven3$estimate, propwseven4$estimate, propwseven5$estimate, propwseven6$estimate, propwseven7$estimate, propwseven8$estimate, propwseven9$estimate, propwseven10$estimate)
range((propwseven_estdiff))
## [1] -0.16398422  0.09655242
median(propwseven_estdiff)
## [1] -0.01813784
#only propi is correlated at 6 months, audio to video
propcsix1 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_prop_op+a_prop_op,conf.int=T, method = "kendall")
propcsix2 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_prop_mom+a_prop_mom,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.933333333333333, 0.96551724137931,
## 0.981481481481482, : Cannot compute exact p-value with ties
propcsix3 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_prop_dad+a_prop_dad,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.0583333333333333, 0, 0, 0,
## 0.0315789473684211, : Cannot compute exact p-value with ties
propcsix4 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_propd+a_propd,conf.int=T, method = "kendall")
propcsix5 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_propi+a_propi,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.00833333333333333,
## 0.0172413793103448, : Cannot compute exact p-value with ties
propcsix6 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_propr+a_propr,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.241666666666667, 0, 0, 0,
## 0.0842105263157895, : Cannot compute exact p-value with ties
propcsix7 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_propq+a_propq,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.233333333333333, 0.448275862068966, :
## Cannot compute exact p-value with ties
propcsix8 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_props+a_props,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.0166666666666667, 0,
## 0.111111111111111, : Cannot compute exact p-value with ties
propcsix9 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_propn+a_propn,conf.int=T, method = "kendall")
propcsix10 <- cor.test(data=subset(sixseven_spreadAV, month=="06"& subj!="17"), ~v_type_token_ratio+a_type_token_ratio,conf.int=T, method = "kendall")

propcsix_ps <- cbind(propcsix1$p.value,propcsix2$p.value,propcsix3$p.value, propcsix4$p.value, propcsix5$p.value, propcsix6$p.value, propcsix7$p.value, propcsix8$p.value, propcsix9$p.value, propcsix10$p.value)
round(p.adjust(propcsix_ps, method = "holm"),3) #only propi
##  [1] 0.554 0.211 0.554 1.000 0.039 0.834 1.000 0.554 0.547 1.000
round(propcsix_ps, 3)
##       [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7]  [,8]  [,9] [,10]
## [1,] 0.082 0.023 0.079 0.359 0.004 0.209 0.691 0.094 0.068 0.492
table(propcsix_ps<.05)
## 
## FALSE  TRUE 
##     8     2
table(p.adjust(propcsix_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     9     1
propcsix5
## 
##  Kendall's rank correlation tau
## 
## data:  v_propi and a_propi
## z = 2.889, p-value = 0.003864
## alternative hypothesis: true tau is not equal to 0
## sample estimates:
##       tau 
## 0.3061568
#most seven props are correlated
propcseven1 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_prop_op+a_prop_op,conf.int=T, method = "kendall")
propcseven2 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_prop_mom+a_prop_mom,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.973684210526316, 1,
## 0.940298507462687, : Cannot compute exact p-value with ties
propcseven3 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_prop_dad+a_prop_dad,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0, 0, 0, 0, 0.045, 0.0103092783505155, :
## Cannot compute exact p-value with ties
propcseven4 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_propd+a_propd,conf.int=T, method = "kendall")
propcseven5 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_propi+a_propi,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.0175438596491228, 0.0612244897959184, :
## Cannot compute exact p-value with ties
propcseven6 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_propr+a_propr,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0, 0, 0.111940298507463, 0, 0.25, 0, 0, :
## Cannot compute exact p-value with ties
propcseven7 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_propq+a_propq,conf.int=T, method = "kendall")
propcseven8 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_props+a_props,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.596491228070175, 0, 0,
## 0.808510638297872, : Cannot compute exact p-value with ties
propcseven9 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_propn+a_propn,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0, 0, 0.156716417910448,
## 0.0567375886524823, : Cannot compute exact p-value with ties
propcseven10 <- cor.test(data=subset(sixseven_spreadAV, month=="07"), ~v_type_token_ratio+a_type_token_ratio,conf.int=T, method = "kendall")
## Warning in cor.test.default(x = c(0.271929824561404, 0.489795918367347, :
## Cannot compute exact p-value with ties
propcseven_ps <- cbind(propcseven1$p.value,propcseven2$p.value,propcseven3$p.value, propcseven4$p.value, propcseven5$p.value, propcseven6$p.value, propcseven7$p.value, propcseven8$p.value, propcseven9$p.value, propcseven10$p.value)
round(p.adjust(propcseven_ps, method = "holm"),3) #2,3,4,6: propmom, propdad, propd, propr
##  [1] 0.740 0.037 0.037 0.035 0.094 0.037 0.175 0.740 0.651 0.094
round(propcseven_ps, 3)
##      [,1]  [,2]  [,3]  [,4]  [,5]  [,6]  [,7] [,8]  [,9] [,10]
## [1,] 0.37 0.004 0.005 0.004 0.016 0.004 0.044 0.42 0.217 0.016
table(propcseven_ps<.05)
## 
## FALSE  TRUE 
##     3     7
table(p.adjust(propcseven_ps, method = "holm")<.05)
## 
## FALSE  TRUE 
##     6     4
propcseven_taus <- cbind(propcseven2$estimate,propcseven3$estimate, propcseven4$estimate, propcseven6$estimate) 
range(propcseven_taus)
## [1] 0.3023256 0.3338215
#are things different by month graph, paired wilcoxon
ggplot(sixseven_basiclevel_home_data_agg, aes(month, r))+ geom_line(aes(group = subj))+
  stat_summary(fun.y=mean, geom="line",aes(group=1), color = "red", size=2)+
  stat_summary(fun.data=mean_cl_boot, aes(group=1), color = "red")+
  facet_wrap(~audio_video, scales = "free")

ggplot(sixseven_spreadAV, aes(v_r, a_r))+ 
  geom_point()+
  stat_smooth()+
  facet_wrap(~month)
## `geom_smooth()` using method = 'loess'

ggplot(sixseven_spreadmonth, aes(six_r, sev_r))+ 
  geom_point()+
  stat_smooth(method = "rlm")+
  facet_wrap(~audio_video)
## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

## Warning: Computation failed in `stat_smooth()`:
## object 'rlm' of mode 'function' was not found

# propvals_med <- sixseven_basiclevel_home_data_agg %>% 
#   dplyr::select(month, audio_video, prop_op, prop_mom, prop_dad, propd, propi, propn, propq, propr, props, type_token_ratio) %>% 
#   group_by(month, audio_video) %>% 
#   #summarise_if(is.numeric, funs(min, max, mean, median))
#   summarise_if(is.numeric, funs(median)) %>% 
#   gather(prop_meas, medianval, prop_op:type_token_ratio)
# 
# propvals_mean <- sixseven_basiclevel_home_data_agg %>% 
#   dplyr::select(month, audio_video, prop_op, prop_mom, prop_dad, propd, propi, propn, propq, propr, props, type_token_ratio) %>% 
#   group_by(month, audio_video) %>% 
#   #summarise_if(is.numeric, funs(min, max, mean, median))
#   summarise_if(is.numeric, funs(mean)) %>% 
#   gather(prop_meas, meanval, prop_op:type_token_ratio)

propvals_long <- sixseven_basiclevel_home_data_agg %>% 
  dplyr::select(month, audio_video, subj, prop_op, prop_mom, prop_dad, propd, propi, propn, propq, propr, props, type_token_ratio) %>% 
  group_by(month, audio_video, subj) %>% 
  gather(prop_meas, propval, prop_op:type_token_ratio)

countvals_long <- sixseven_basiclevel_home_data_agg %>% 
  dplyr::select(month, audio_video, subj, y_op, MOT, FAT, d, i, n, q, r, s, numtypes, numtokens, numspeakers) %>% 
  group_by(month, audio_video, subj) %>% 
  gather(count_meas, countval, y_op:numspeakers)


# ggplot(propvals_med, aes(fill = prop_meas, linetype=audio_video,prop_meas, medianval))+
#   facet_wrap(month~audio_video,ncol=1)+geom_bar(stat="identity", color = "black")
# ggplot(propvals_med, aes(fill = prop_meas, linetype=audio_video, month, medianval))+
#   geom_bar(stat="identity", color = "black", position ="dodge")

# ggplot(propvals_mean, aes(fill = prop_meas, linetype=audio_video, month, meanval))+
#   geom_bar(stat="identity", color = "black", position ="dodge")
ggplot(propvals_long, aes(fill = prop_meas, linetype =audio_video, month, propval))+
  stat_summary(fun.y=mean, geom="bar", position = "dodge", aes(linetype=audio_video), color = "black")+
  stat_summary(fun.data=mean_cl_boot, geom="pointrange", position=position_dodge(width=.9), 
               aes(group = interaction(prop_meas,month,audio_video)))

ggplot(countvals_long, aes(fill = count_meas, linetype =audio_video, month, countval))+
  stat_summary(fun.y=mean, geom="bar", position = "dodge", aes(linetype=audio_video), color = "black")+
  stat_summary(fun.data=mean_cl_boot, geom="pointrange", position=position_dodge(width=.9), 
               aes(group = interaction(count_meas,month,audio_video)))+
  facet_wrap(count_meas~audio_video, scales="free_y", nrow=6)